The present invention relates to an image processing method and an image processing device, and more particularly to an image processing method by which each pixel of gray-scale data obtained by a raster scanning is determined to either include or not include an edge of a picture, and to an image processing device in which gray-scale data obtained by a raster scanning is converted into binary image data.
In a picture scanning apparatus such as a facsimile machine, various image processing operations are performed to scanned image data so as to improve an output picture quality.
For original printed characters, since intensity difference between black portions corresponding to characters and white portions corresponding to blank spaces is very large, a process which enhances the contrast of the black portions and the white portions has been realized. As an example of such a process, a Modulation Transfer Function (MTF) correction process has been applied so as to eliminate a deterioration of a picture such as blurring, which occurs in an optical scanning system.
Additionally, when scanning a half-tone image such as a photograph, gradation of an original image is represented in a reproduced image by binarizing gray-scale image information in accordance with a threshold value obtained by means of the systematic dither matrix or by using a half-tone binarizing process such as the error diffusion process in which the dispersion of intensity caused by binarizing is diffused to surrounding pixels.
When the original image includes only characters (hereinafter called non-half-tone image) or only pictures having half-tone image, the above-mentioned processes can be applied to improve the reproduced image. However, when a non-half-tone image and a half-tone image are mixed in an original, the following disadvantages result.
That is, if the MTF correction is applied to a half-tone image, gradation is not maintained and thus the half-tone image quality is deteriorated. On the other hand, if the half-tone image process is applied to a non-half-tone image, boundary portions (edge portions) lying between black portions and white portions are blurred.
In order to eliminate the above-mentioned problem, an image processing device in which the MTF correction process or the half-tone image process is selectively applied is realized. In this device, the MTF correction process is applied to an image portion determined as a non-half-tone image, and the half-tone binarizing process is applied to an image portion determined as a half-tone image.
However, the above-mentioned device requires an image area separation process, a non-half-tone binarizing process and a half-tone binarizing process, and thus manufacturing costs thereof are high.
A device eliminating such a problem has been realized, such as a device in which the half-tone binarizing process is applied for basically all pixels excluding edge pixels. An edge pixel is a pixel corresponding to the transition of a white image to a black image. A pixel is determined to be such an edge pixel by comparing the pixel being processed with bordering pixels. A pixel is determined to be an edge pixel when intensity difference between the pixel and bordering pixels exceeds a predetermined value. A black pixel signal is output for an edge pixel.
In this device, a non-half-tone image such as a character positioned in a half-tone image area can be reproduced clearly, and a high quality picture can thus be obtained at a relatively low cost.
However, the above-mentioned conventional device has the following problems. That is, while enhancing a character in a white area can be performed effectively, a moire generated when scanning a mesh picture is also undesirably enhanced. Additionally, when scanning an original whose surface is not white, improvement of a reproduced image obtained is not sufficient.